PHOLADID.E. 97 



of the zeal, ability, and conscientious labour with which he 

 has investigated the subject; and he may be justly termed 

 the apostle of this theory. His researches have been 

 carried on_, with scarcely any intermission, for more than 

 twenty years ; and although I have ventured to disagree 

 with him in the present instance, I still entertain a pro- 

 found respect for his opinion. Were I to become con- 

 verted, it would be solely by his arguments. The careful 

 and precise experiments made by him leave no doubt 

 that the shell can be used by man as an instrument of 

 perforation ; it by no means follows that it is so used 

 by the mollusk. It is easy to scrape with the edge of a 

 limpet-shell a cavity in chalk or shale, such as Patella oc- 

 cupies ; but can it be imagined that in this case the shell, 

 instead of the foot, is naturally employed for that pur- 

 pose? I believe that all the phenomena which have 

 been attributed by Cailliaud to the mechanical action of 

 the shell may be accounted for by the theory of Sellius, 

 or rather of his predecessor Rea^mur. For instance, 

 the fine and regular striae, which are observable on the 

 sides of the cell of a Pholas, are unquestionably caused 

 by the friction of the spinous ridges that ornament the 

 shell. These strise are wanting at the bottom of the 

 cell, and are replaced there by a far more delicate elabo- 

 ration, which I am of opinion is produced by the sucker- 

 like motion of the foot. Assuming the latter to be the 

 instrument of perforation, the shell would partake of its 

 motion, and would rasp the walls of the cell while the 

 foot was doing the work of excavation. The prickly 

 surface of the shell, stretched to its full extent by the 

 adductor muscle, is pressed against the sides or walls of 

 the cell, and acts as a fulcrum. Born's idea, repeated 

 by Cailliaud, viz. that the foot acts as the fulcrum or point 

 of leverage, is exactly the reverse of mine. Besides, let 

 VOL. III. r 



